Podcast
The CIO Journey: From Information to Innovation in Modern Healthcare
Bill Russell
Founder of This Week Health
There is a clear distinction between individuals who seek to promote themselves by advancing technology in healthcare and those that seek to move healthcare forward by advancing meaningful technology. Bill Russell is undoubtedly the latter. Formerly the CIO for St. Joseph’s Health, Bill oversaw a $6.5 billion healthcare system with sixteen hospitals, and he has contributed to IT advancements such as agile cloud development in healthcare. On this episode of Healthcare Market Matrix, Bill joins host John Farkas to discuss the unique challenges CIOs face, adopting healthtech solutions, AI’s impact on healthcare, and much more.
Listen Now
Transcript
Introducing Bill Russell
John Farkas:
Well, welcome everybody to Healthcare Market Matrix and I’m really excited about who we have the chance to talk with today. There are people who seek to promote themselves by advancing technology and healthcare and then there are people who seek to move healthcare forward by advancing meaningful technology. And today, I know we have the chance to talk to the latter. Bill Russell needs no introduction for many, but for the few of you who don’t know, he is the CEO and founder of Health Lyrics and the podcast host of This Week Health.
Bill has served on lots of executive teams in healthcare, higher education and Fortune 500 consulting practices. He’s the former CIO for St. Joseph’s Health, overseeing a 6.5 billion dollar system with 16 hospitals. He has been in the situation to accelerate IT adoption and strategies and methods including agile cloud development and has successfully invested in a number of startups driving lots of innovation in the healthcare sector and is currently involved in executive coaching and advisory services. He’s providing technology and future focus coaching to a number of health leaders and offers advisory services to the healthcare providers and tech startups and investors through Health Lyrics, which I know we’ll talk a little bit more about. Bill, welcome to Healthcare Market Matrix.
Bill Russell:
John, thanks for inviting me. And anytime you hear your introduction, it makes you feel old. And listening to my introduction, I’m like, “Wow, it seems like forever ago that I was a CIO for St. Joe’s,” and some of the other stuff you didn’t mention. It is 20 plus years in consulting prior to that. And over time, you just get a different perspective on things, I believe.
John Farkas:
Well, and that’s part of why you’re here because you’ve lived to tell the tales. There’s been a lot in that tenure. But really grateful for you joining us here today. And I know that you’ve been in a number, like we mentioned, a number of leadership positions over your career. But as we’ve heard you talk and mentioned a number of times, the CIO role in healthcare has been one of, if not, the hardest. And I’m just curious from your perspective why that is. What do you see as the challenge in that seat?
Bill Russell:
It’s interesting because I’ve had some really hard jobs. I’ve been over a service organization for a Fortune 500 company. Had 2000 engineers reporting it to me in 42 states, I believe, managing 200 million in consulting revenue and that kind of stuff. And somebody would say, “That’s hard.” And then we took that company through chapter 11. “Oh, that’s hard.” And I’m here to tell you they pale in comparison to being a CIO at a health system. And I think the best way to think about it is the CIO role in healthcare is really the CEO of applying technology to that healthcare system. Okay? And what I mean by that is you play every role. Sometimes you set the table and you bring the technology in, sometimes you implement, sometimes you have to rally a group of people around a concept or an idea or around a problem set that needs to be solved.
Sometimes you’re an advisor and counselor. You have to have vision, ask the right amount of questions of people. You have to get the right amount of money to make sure you’re managing your IT organization well. And then of course, then you have all the people challenges, managing people and the various things that go on. And then I think the other thing that adds the complexity because people might say, “That’s not that hard,” and it is hard, but the complexity is healthcare. And when people say the word healthcare to me, I hear a lot of different things. And when a lot of people say the word healthcare, they’re like healthcare provider. Well, to me, a healthcare provider is six types of different organizations. It can be a private equity or venture capital backed organization. It can be an academic medical center and everything in between.
And each one of those is made up of 40 businesses that sometimes are tied together really well and sometimes are not tied together that well and in some cases have competing priorities. And so I set that up. And then I say, when I first went into healthcare as the CIO, it wasn’t uncommon for me to have a meeting at 10:00 AM to talk about surgery. At noon, I’m talking about oncology. At one o’clock, talking about supply chain. Two o’clock, I’m meeting with the vendor and talking enterprise architecture. Three o’clock, I’m talking food services because we have food services in every one of our hospitals. And at 4:00 PM, I’m sitting down with a CFO for a multi-billion dollar organization and we’re talking budget and finance. So hopefully, I painted the picture for why I believe it’s one of the hardest roles, period, out there.
John Farkas:
Yeah. And needing to have some level of mastery and operational understanding of what goes into each of those, right? Because you’re looking to support them and understand the issues and challenges faced by each one of those segments.
Bill Russell:
Yeah, I joke around that when I first went into healthcare, I came from a technology background. I didn’t come from a healthcare background because that’s what our health system wanted. They wanted some new thinking, but I had to get my admin put together a list of all the different practices because I was sitting across from a doctor of some kind and they would say, “Well, I practiced this medicine,” and I would’ve no idea what that was. And so for the first couple of weeks, I had cards and I memorized all the different practices within the health system just to know who I was talking to and what they did.
Tips for Pitching to CIOs
John Farkas:
So good lead in to one of the questions I have here, and we have a number, but I’m just curious. Our audience and who’s listening to this podcast are marketing leaders of health tech organizations. So if you just heard what Bill said and the CIO is a target, somebody you’re trying to get the attention of, that puts forward a pretty significant mandate because first of all, you are working as the CIO to balance an incredibly complex budget with lots of competing demands and needing to make decisions based on clear value. And so if you were to coach a marketing leader of a technology solution selling into a health system, how would you help them get their message across? What would you say is the thing that they need to make sure they’re doing a great job of, if they’re going to get the attention of a CIO and help them understand what it is they’re bringing?
Bill Russell:
It’s so important to know each health system and what their priorities are. It’s interesting, when I started the podcast, we would interview CIOs and there was a few sales managers that told me your show is required listening for our sales organization. I was like, “Really? Tell me about that.” They’re like, “Look, you’re having the conversations that many of us can’t have and you’re asking the questions that many of us can’t ask of the CIOs.” And by the way, I’m not the only one doing it now. You can go to Becker’s, you can go to Health System CIO, a bunch of us interview the CIOs and you can get the priorities from them. So the first thing is understand their priorities. The second is you have to have a story and your story has to have a … What’s the best way to say it? A demonstrable return on not only finance. We all know that. All needs a financial ROI, but it needs a demonstrable return on the resource in investment of moving something forward.
Because, by the way, finance and resources aren’t the only two things I’m going to invest in your project. The other is I’m going to invest my relational capital, my political capital that I’ve built up. If I’m getting behind a project, I’m essentially investing whatever relational and political capital. Because I’m going to have to pull in some doctors and get them excited about it or pull in some administrators and get them excited about it. So I’ve built up some capital. I’m spending that capital when I bring you in, you have to have a story that helps me to make that case internally. No matter what it is. And it has to be clear. It’s amazing to me the number of times where I’m sitting across from somebody and they’re telling me, “You have to implement our system. It’s the blah, blah, blah.”
And I say, “Well, tell me what the return is from a financial standpoint,” and it’s murky at best. And I’m like, “All right, you should have that down.” In fact, in today’s economy and financial situations for health systems today, that has to be rock solid. You have to be able to say, “We can save you.” And it can be in actual dollars, can be in reimbursements, could be in revenue capture, it can be in hours. A lot of times now we’re talking inefficiency, you’re saving us hours. It can be in any of those things, but that story has to be very solid. “You’re going to invest this and you’re going to get this return.” But the other thing is really understanding the problems. So nurse staffing shortages and nurse staffing, just clinical burnout in general. These are real problems. What is your story to go against that and how is that delivering.
The financial burden that I mentioned earlier, what is your story and how are you participating in that? Security, PHI, PII vulnerabilities, how are you participating in that? That’s just to name a few and just understand, there’s an awful lot of things going on in the health system because I’ve had salespeople look at me and go, “Hey, you know what? Bill, I don’t understand why this health system’s not doing this.” And I’m like, “Let me tell you what some of their priorities are.” And I’ll rattle them off and I’ll say, “They have to address sepsis.” And they’ll go, “What is sepsis?” I’m like, “Well, it’s leading cause of death in a hospital. So they have to address that. They have to improve quality.”
They have a staffing issue that’s going on. They have unrest amongst their clinicians, they have burnout issues and those kind of things. They just had to do a reduction in their IT staff. Every one of their vendors is coming to them with a 10 to 15% increase in their price for ongoing software contracts. And you’re asking me the question, are you really asking me the question, why aren’t they prioritizing the stuff that you’re talking about? And the answer is, you might be number 10 on their list and they’re not going to get to number 10 this year.
John Farkas:
Yeah, it’s a really good point. That clarity to value has to be there. It is so frequent that we see organizations wanting to talk all about their technology in some way. Here’s our really cool shiny object that does X, Y, Z, which is great. And if you don’t have a really clear line to value, it means virtually nothing. And that’s a such an important element.
Bill Russell:
By the way, help me sell it internally. If I have to build out all this stuff myself, I don’t have the time and it’s going to be really hard. I want you to hand me this stuff and say, “Hey, if you’re talking to a clinician, here’s the story, take this story to them.”
John Farkas:
Yeah, that’s a great point. Have the internal campaign framework set out so that you can help your advocate bring it forward. That’s really good.
What Problems Are CIOs Solving Today?
John Farkas:
So you outlined several of the problems or you listed several of them, but what are some of the few top problems you’re seeing come across CIO’s desks right now that they’re willing to or feeling urgent about dedicating some resources to solving?
Bill Russell:
The number one problem, and you may not hear this from CIOs because none of them want to really talk about this overtly because it’s a difficult thing, but the number one problem is the financial hole that healthcare is currently trying to climb out of. And we’re seeing some of the leaders climb out of it. You’re seeing Intermountain and Mayo and some of the others climb out of it, but some of the others are really struggling to get out of that. That’s the number one thing that’s happening. We come through the pandemic, which was a jolt to the system for sure. Now we have inflation, we have a sluggish economy, which is never good for healthcare. And oh, by the way, while that’s all happening, the competition has only grown for imaging, for a lot of the things where we used to make money-
John Farkas:
That were revenue sources.
Bill Russell:
… for primary care. So that’s the environment. So that’s number one. So they’re looking for efficiency and potentially revenue generation, so technology that helps in generating revenue, any kind of tech, this is why we’re talking so much about AI right now because there’s the belief that AI can start to step in and take some of these mundane, but some of the mundane work that we all do including in IT and in the clinical side and start to automate some of that. I would say it goes beyond that too though. I mean, once you get past the financial access is always something that people are looking for.
How do we give people more access to healthcare, remote patient monitoring, patient flow? I’m hearing a lot of command centers be stood up because we would have people get stuck in the system and instead of moving efficiently through, they would get stuck. And even to the point of discharge, we have some patients who are sitting maybe four or five hours taking up a room in a hospital where they could have been discharged five hours ago. And I mean, all those things represent assets and I think that the last one I would mention is digital expectations have changed and it’s due to all the things that people talk about, it’s due to the Amazon effect or-
John Farkas:
And so when you say digital expectations, you’re saying how can people engage digitally in whatever framework?
Bill Russell:
Right. You hear this all the time and maybe you’ve experienced it yourself, which is, “Hey, why can’t I get an appointment through my phone? Why can’t I cancel it? Why can’t I reschedule it? Why can’t I get my discharge instructions via some digital tool that I can then show to people on my TV at home?” Our expectations have changed because every other industry has changed them because we’re already in retail, we’re there, everywhere else, we’re there. And it’s almost cliche to talk about now what we’re pushing the CIOs in terms of the conversation is if those are their expectations, what are the projects we’re putting in place to fulfill the expectations that they do have in terms of how healthcare will operate over the next five years?
Finding New Tech Solutions through Other Health Systems
John Farkas:
Yeah, that makes a lot of sense. So thinking back at your days at St. Joseph, if you were looking to or how were different vendors, different technology organizations, how are they coming across your desk? How are you becoming aware of them? What were some of the channels and methods that you were hearing of new opportunities?
Bill Russell:
I ended up writing an article about this. The number one way that people hear about solutions is from other health systems period. And it’s like a three to one, four to one, five to one margin over whatever number two is. Because we just had one of these events and the CIOs were sitting around and somebody said, “Hey, we were able to get more efficiency out of our MRI machines by utilizing this thing.” And three people swarmed that person after that presentation to say, “Hey, I needed more information on that.” That’s a very common approach. You’ll see that happen at conferences. You’ll see people say, “Hey, we’re kicking off a project here,” and we think they’re going to start at Gartner, or we think or they’re going to start a class or one of those other services. But a lot of times what they do is they’ll essentially go out into the field.
There are groups of people where they can send an email to and say, “Hey, who’s doing this specific thing for hospital at home with remote patient monitoring?” And they’ll get two or three responses. They’ll get their teams on the phone with them and that’s how they’ll learn. That’s number one. Every now and then, I would do the booth crawl at various conferences, but most of the time when I go to do the booth crawl, I’m looking for my existing partners so that I can have conversations with them and whatnot. Rarely was I doing the booth crawl to find a new solution. Every now and then I’d have a doctor with me and the doctor would say, “Hey, I want you to see this thing, or I want you to take a look at this thing.” That would happen.
And then there’s a story now that has sort of grown legend in that one of the people who called on me used to be with Salesforce and he now does sales coaching and he has this whole section where he talks about effectively marketing to a health system and I’m the muse in this whole thing. He sent me emails day in, day out. It felt like that. He said he sent me seven emails before he got a response and it led to a seven figure deal for him and for Salesforce when he finally got through to me. But again, that requires such persistence and he was really good at identifying the things that I was thinking about and just saying, “Hey, I heard this. I heard you do an interview, I heard you speak.” And he would drop these things and eventually, he got in touch with me. We got into a room and he listened and found a way to help us.
John Farkas:
So some of what I’m hearing there that I think is really important is, first of all, you have to know what’s going on in the organization. You have to know what’s top of mind. You have to understand what are the core problems that they’re facing in trying to solve and tailor whatever approach you might be taking with some very clear understanding of the context. Because if you come out in blind and you’re just throwing a blanket solution or a blanket approach, it’s likely not going to get the attention. The other thing is recognizing that peer to peer is going to be some of the most effective channels. So figuring out how to use existing deployments, existing successful deployments and amplified those out so people can see that and understand what in fact is going on, what in fact is being realized as far as value goes in a real respectable system that people have connection to. That seems like those are some important takeaways.
Bill on Failed Implementations
Bill Russell:
Yeah. You cannot afford to have a failed implementation. It’s too small. It’s too small of a community to have failed implementations. If you do, I would go back and remedy those as quickly as you possibly can.
John Farkas:
Yeah, talk about what you’ve experienced in that regard, because I’ve definitely heard, especially recently and especially around the AI universe of some things not going well and that spreading pretty rampantly through the rumor mill.
Bill Russell:
Yeah, I mean, I’ll speak to the rumor mill. The rumor mill is pretty obvious. It’s one of those where you’re sitting with a group of CIOs and we get into that old curmudgeonly kind of thing. “You think your life’s bad. Let me tell you how bad mine is. I went to implement X, Y, Z technology. Oh my gosh, these people were idiots. They absolutely didn’t know what they were doing and blah, blah.” Even if that company did know what they were doing or whatever, that person who’s relaying that story isn’t going to say it that way. They’re going to tell you all the things that went wrong. They’re going to embellish in some cases. I mean, you cannot afford, it just happens. I mean, I’ve just seen it, I’ve seen it, I’ve heard it over and over again. So you have to make sure you remedy those as quickly as you can.
At the minimum, you want to get people to be neutral on you if you have a failed implementation because we know. In healthcare, we know that a lot of times we’re the problem. It took us six months to get the contract together. We didn’t get you the data correctly. We didn’t get you the integrations correctly. Our doctors didn’t engage. There’s any number of reasons the project didn’t go well that are literally our fault. But if it’s going to fail, at least get that neutral feeling of, “Hey, let’s pull the plug on this. It doesn’t seem like this is a priority for you right now. Let’s get to a point where it’s stable and we can cut our losses or whatever.” That’s the minimum. At most, I would tell organizations, “Man, you have to be really good at managing your clients through a project. You should know what it takes, who needs to be involved.”
In fact, I was so impressed when I had a solution and they came in and said, “These are the people that we are going to need to engage with. This is the amount of time we’re going to need from those people. This is the connections we’re going to need. Here’s the security framework that we’re going to utilize.” They knew the questions we were going to ask before we asked them and they provided them complete with, “Here’s a template for the project management.” A lot of them will provide former healthcare staff that have been through it before that can help us to walk through.
John Farkas:
We can walk alongside. Yeah.
Bill Russell:
All that stuff is really important. And by the way, I mean, EPIC has been successful really for a lot of reasons. But the two early on I would say is, one is they were the only system with an acute and ambulatory integrated system when meaningful use came out. So that’s the number one reason they won is that. But the number two reason they won is when EPIC comes into your healthcare system, they say, “You’re going to get this many doctors trained. You’re going to do this, you’re going to do this. Here are the recipes for success and we’re going to be with you and walk you through it every step of the way.” And healthcare, they like that. They’ve grown used to it. They’ve grown used to it with EPIC and they would like to see it from more players who come in there and really know how to get stuff done in healthcare.
John Farkas:
Which is really the challenge for a lot of early stage health tech companies, right? Because they have a relatively thin team. Sometimes they don’t have that savvy expertise. And so spending the time on the front end to get that, because you can’t afford a failed implementation, especially early on, you have to come … And so spending the calories that you need to go to school and understand what it’s going to take to be successful and being successful at almost any expense, especially in the early days, is a really critical step. Because, first of all, you need the experience and you’re going to continue to need the experience because technology isn’t enough. If you’re a CIO and you’re implementing a six figure or a seven figure, an eight figure system and you can’t afford for it to fail. Your reputation can’t sustain it and your budget can’t sustain it and the system can’t sustain it right now.
Partnering with Health Systems
Bill Russell:
Really early on, I will say it’s so important to get a partner and I tell really early stage startups, “Set aside some ownership for a partner.” So they’re going to partner with me at St. Joe’s. What they should do is offer me warrants. They should offer me some ownership. And the reason they should do that is because I’m taking a flyer on them. They’re so early on and what they need from me is they need some of my doctor’s time to say, “Man, I see what you’re doing here. You shouldn’t do it this way. You should consider doing it this and this because you don’t really understand how this works in healthcare.”
And they might need some help from my security team to shore up their security framework or my integration team to do some things. They want me to feel like I’m an owner so that I’m going to give them as many resources as possible. I’m also going to be in senate to give them a positive review out into the world. So I would say set aside some ownership early on for one or two players and then I would say stop giving away ownership once you have three or four clients. I mean, I see people who are like, “Oh yeah, we gave them a piece of ownership and they’re on client number 10.” I’m like, “What are you doing? Your product should be baked by now.”
John Farkas:
Well, having a very clear principle approach to that onboarding in those … You need to go into those first deployments, understanding that we’ve got to get the system together. That’s our purpose right now is to get our system together so that by the time you get to number four, you’ve got it down and you have the opportunity to push it forward.
Bill Russell:
And hopefully in working with me, I’m going to give you the project plan that we used and then you could use that with other clients.
Bill’s Expectations for AI’s Impact on Healthcare
John Farkas:
Yep. Yep. Good. That’s a great word. So let’s shift gears a little bit. I know you’re the host of This Week Health, we hear you talking a lot with CIOs and in that community. We know that AI is a strong emerging force on a number of fronts. What are you seeing some of the opportunities? If you look at the opportunities that AI is going to open up for healthcare and how it intersects with the CIO universe, what are you seeing as some of the big beachhead there that are going to be transformative in the next few years?
Bill Russell:
AI is going to change everything and it’s going to do so very rapidly. And what I’m telling CIOs right now is stop putting up roadblocks. Stop moving so slowly because this is going to change everything and your competitors are trying to figure out how to use it. One of the CIOs I respect greatly is Mike Pfeffer from Stanford and they’re already using AI to draft notes for doctor’s responses and then once it’s drafted, the doctor can look at it and say, “Go ahead and send it.” And they’re saving a certain amount of time and the studies will continue to be done. But those doctors used to have to write those notes and they actually did a study on the notes for empathy and for completeness, and the AI generated notes were more empathetic and more complete than the ones the doctors were doing. Why is that? Well, the doctor just worked an eight to 10 hour day.
John Farkas:
They don’t have time.
Bill Russell:
Right. They’re at the end of their day. They’re just like, “Yeah, go ahead and take the medication or let me do it.” Whereas AI sitting there going, “Hey, thank you for your question. I really appreciate that. I’ve been looking through your record,” all the stuff that the doctor wouldn’t type. Anyway, I start with saying it’s going to change everything. AI is going to be an information assistant. It’s going to follow me all day. It’s going to document what needs to be documented in my day. It’s going to send emails or at least draft emails for me that need to be sent. It’s going to assist me with tasks that can be done digitally, like ordering something or those kind of things. It’s going to do research on my behalf and it’s going to provide me all sorts of support in my work.
This is what AI has the potential to do. Clinically, this means that every physician will have an AI assistant that can gather the information for their patients that they’re going to see that day, all right? And it’s going to get the latest research from the journals on those patients that they’re going to see. It’s going to score those for relevance for them. It’s going to provide them a summary on some of the bullet points for the patients that they’re going to see. It’s going to listen in on the patient visit and it’s going to document that patient visit. It may even pull up some relevant information from the EHR while they’re having that visit. You get the picture. From a clinician standpoint, the anticipation is this is going to be used everywhere to do things that we used to have to go to our mouse, go to our keyboard and go, “I wonder what the last 10 readings were on blood glucose for this patient.”
And by the way, I’m just talking about clinicians now. Nurses have a lot of, again, tasks that are, depending on their level of certification, some of the tasks that nurses are asked to do are quite frankly beneath their level of certification. And those tasks we have proven now can be done with cameras and computer vision utilizing AI and those kind of things. What we are really cognizant of right now though, is AI washing. Look, we’ve been doing predictive models for years without algorithms. That is not AI. AI is with every piece of information that washes over the AI algorithms that are there, it actually gets more intuitive. It gets smarter. It knows where to find information better. It knows how I function better. It gets smarter as we go along. That’s what we’re looking for is algorithms that get better with use.
John Farkas:
And clearly, one of the things that … what we know right now is one of the biggest needs that is facing healthcare has to do with the clinician shortage and what is going on. Really critical that whatever solution you’re bringing forward has a very clear line to actually helping and lifting the burdens off of clinicians. It’s one thing to throw another technology layer over something and it’s another thing to have a solution that is, from its core, from its foundation, built to save cycles and to improve care quality. And I think that that’s a real critical element as you consider how you are messaging, how you are bringing these ideas forward, how’s that going to be clearly related?
Bill Russell:
Yeah. It goes back to what we talked about earlier. You have to have that demonstrable … I mean, this is why you need that early client as well so that you can write that paper that says, “Hey, we reduced their clinician time in notes or follow up or in documentation by 25%, which it allowed them to do whatever they want with that time.” But in a lot of cases, the clinicians, they say they’re suffering from burnout, but when we give them more time what they do? They see two more patients a day. Go figure.
The Need for AI Policies in Healthcare
John Farkas:
Bill, as you’re interacting with CIOs and in the context of AI, where is the biggest knowledge gap right now? What’s the gap that needs to be closed between the average? I know there’s a huge spectrum of embrace right now around AI ideas and there’s been a whole lot that’s just come out in the last eight months around what is now possible as the large language models have come to roost. What are you seeing as the gap that exists in fundamental understanding right now that needs to get closed that organizations who are bringing AI solutions forward would do well to invest in helping people understand?
Bill Russell:
This is an interesting question and part of the discussion we had this past weekend.I hope people understand, you just can’t pop technology in a place like healthcare. You can do it in some industries. You just can’t do it in healthcare. It’s highly regulated.
John Farkas:
It’s too much context.
Bill Russell:
Right. It’s highly regulated. And the mistake could cost someone their life or the life of a loved one. So either way, you can’t just pop it in there. There’s foundational work that needs to be done before AI can be widely adopted and used. I would start with we have to create policies, and I asked this group of CIOs, how many of you have an AI usage policy in place? And of the 13 that were in the room, two thirds were working on it, but no one had one done.
John Farkas:
Wow. Yeah.
Bill Russell:
I know it didn’t, but it’s like all of a sudden with chatGPT, all of a sudden became, “Oh my gosh, we need an AI policy.” Now we were using AI prior to that, but it wasn’t as pervasive and there wasn’t as much demand to use it. And so we didn’t put the policies around it and we have to put those policies around it. The policies have to protect people, data has to be protected. Obviously, we’re doing AI in supply chain, we’re doing it in IT, areas where there’s no lives at risk, but we have to do that.
And while we’re talking about policy, this week’s poll, I asked … So I do polls every Monday on LinkedIn just to get the voice of the industry. And my question this week was around policy. If you went to a healthcare system and you read their AI policy, what would you be really concerned if you didn’t see on there? And with LinkedIn, you can only give four options. I would’ve given eight if I could, but the four options I gave were data privacy and security, transparency and explainability, fairness and bias mitigation and human oversight and accountability. These are just four of the headings. I think that should be in every AI policy.
Data privacy and security, transparency, explainability, we can explain why the AI algorithm is telling us what it’s telling us. Fairness and bias mitigation because we know that these algorithms were trained and they’re trained on data from the internet. And that isn’t necessarily bias free information. And the example I give people is we don’t know if chatGPT has been trained on CNN or FOX News and if it was trained on one or the other, it’s going to answer the question different. And so fairness and bias mitigation or human oversight and accountability.
And right now within healthcare, I see in what I’m recommending for people is the human needs to be still the connection to the patient. AI can’t automatically do anything yet. It can reveal information, it can bring it up to the forefront, it can draft things, but the person who actually read it is a human. It’s a physician, it’s a nurse, it’s somebody who can say, “Because I’m pressing this button-“
John Farkas:
I’m responsible.
Bill Russell:
“… I’m responsible for this.” Let me give you the results so far. Data privacy and security, 44% people said, “Hey, if I don’t see that in the policy, I’m going to be concerned.” And 40% said human oversight and accountability. So there’s 84% right there think that those are probably the top two topics that need to be addressed in any AI policy right now.
John Farkas:
Very, very insightful. I know we’re working with a group right now that is very actively working to promote the importance and the methodology for doing an audit of large language models to help ensure if you’re deploying any form a large language model, you better know what’s in there and you better know where it’s coming from and be able to look at what’s going on. And that’s going to be an increasingly important element as this stuff comes forward. And the whole privacy framework around it obviously is essential. What do you see anybody currently doing work on helping promote guidance for those policies other than yourself
Bill Russell:
And I’m. I haven’t had a consulting client in six months. I’m so busy with the stuff that I am doing, building the community and stuff over here. I will give you one funny answer. I said, “How many of you written a policy?” And one of the CIOs said, “Oddly enough, we asked chatGPT to write a AI governance policy for healthcare. And it wasn’t half bad. I mean, it was pretty good actually,” is what the person said. And I’m like, “That’s interesting. AI writing a policy, our large language model writing a policy to oversee AI and implementations.” But quite frankly, based on my usage of chatGPT and GPT 4, it probably would write a fairly good policy statement. Who’s advising on this? At my health system, it would’ve been compliance. We would’ve had physicians. We would’ve had our legal team. Quite frankly, I’m recommending that everybody set up an AI governance board. It’s going to be pervasive in healthcare over the next five years.
It’ll be slow to get going but it will pick up steam very rapidly. And the same way you have governance around data and governance around applications, you’re going to need AI governance because you said transparency in the model. There’s no transparency in the chatGPT model. Nobody can tell you what it’s been trained on. I mean, they can generally tell you what it’s been trained on, but outside of that, we don’t really know what it’s been trained on. And so if it doesn’t have that level of transparency, should it be used in healthcare? Some people are saying yes, as long as it has human oversight, but my encouragement to any healthcare system listening to this is the same way you have an ethics board and those kind of things, you have to determine how this is going to be used in your system and what’s appropriate. Keep in mind if you block it and slow it down, you might be signing your own foreclosure notice. I mean, because it can fundamentally change the economics of healthcare. And if it does that and you decide to step back, don’t be surprised if the financial-
John Farkas:
Don’t be surprised if it bites you.
Bill Russell:
Right.
AI’s Impact on Other Industries
John Farkas:
Yeah, that makes sense. Well, that’s definitely good input. I know that you recently shared an article on LinkedIn where Scott Galloway shares how the economic impact of AI won’t be the catastrophe that many predict. Where do you see that? Where do you see the biggest improvement? Where do you see the biggest impact on AI?
Bill Russell:
We’ve already talked about somewhat in healthcare where I see it, if I were to say where we’re going to see the biggest impact upfront the next two years, we’re going to see it in finance. Imagine AI models essentially managing your portfolio. Now, instead of trusting that financial planner and whatnot, you’re really trusting an algorithm that is looking at way more data points than that flashy person who sold you on, “Trust me with your money, I’ll take care of you.” I think we’re already seeing a lot of AI models in finance. Finance will probably lead the way in this retail will be a close second. When I say close second, I mean, hours behind finance in terms of the implementation of AI. We’re going to see AI really hyper personalizing our experience in retail in a way that is almost unimaginable right now.
And then what will happen, and I had this conversation with Taylor Davis who originally put me onto this concept. Taylor Davis was with class and he led the arch collaborative for many years for class. He was a VP over there and he just started his own deal. And when we were talking about this, he said, “We’ll see industries literally change in what will feel like overnight over the next couple of years. And then healthcare will look at it and go, ‘Oh my gosh, this really is fundamentally going to change things.’ And then we will start adopting it.” So it still might feel really slow.
This thing started, I’m saying the start date was GPT 3 and 3.5 just before Christmas of last year. Just before Christmas of 2022 is around the timeframe that this started. I will say that’s when the clock started. Within three years, we’ll be looking at this across the board in healthcare. It’ll be in supply chain, it’ll be in revenue cycle, it’ll be in IT. It’ll be pervasive in those areas. Administration, it’ll be pervasive in that area. And then you’ll start to really see it take off in the clinical side by then. And it’ll be pretty well expected. A doctor coming out of med school will be like, “You don’t have any AI tools that are going to help me be a better physician? I’m going to go to a different health system.”
John Farkas:
Gotcha. Yeah, that makes total sense. It is going to be a day in this next little bit and how we embrace the waves that are coming is going to make or break a lot of these organizations, for sure.
Why Healthcare Needs to Integrate Quickly
Bill Russell:
John, can I give you my disclaimer? My disclaimer is in all the years I’ve been in healthcare, which is now a little over a decade, all the changes that I thought would be two years away have been four. And the ones I thought would be three years away have been six. We moved to the cloud in 2012 at St. Joe’s, 2012, 2013, and I’m like, “Everybody’s going to move to the cloud.” When they’re moving to the cloud, they just are now moving to the cloud. It’s a decade later from when I thought. I can’t believe this is going to be widespread adoption. It took almost a decade.
John Farkas:
Only a decade.
Bill Russell:
Just moves a little slower. It’s a little …
John Farkas:
What I’d like to hope and is that the economic incentives right now are so undeniable that the innovation may come quicker. I hope they can see their way.
Bill Russell:
That’s why I’m saying that. I do believe that the economics of healthcare are creating a sense of urgency. I believe that the labor situation in healthcare is creating a sense of urgency. I’ve heard people say we will never have enough doctors or nurses again in our lifetime. And so if that’s going to be the case, and that is … We’re not trying to put doctors or nurses out of work. We’re trying to make sure we have enough care knowledge and caregiving being done. And that can be done through a really smart chatbot that’s been trained on the right information and with the right oversight and the right transparency around that model. There’s a belief that 50% of all primary care could potentially be at least triaged if not provided, outright provided by AI within the next five years.
John Farkas:
Yeah. And that scares a lot of people. And if you look at the numbers, it’s almost necessitated, right? I mean, it’s got to change and we’ve got to be open to new ways of addressing some of these things and it’s a big wide open door for really smart technology. But boy, it has to be well-founded.
Bill Russell:
And John, you were worries I wasn’t going to say anything provocative.
John Farkas:
There you go. I think that you jumped right out there, right in it. But anybody who’s looked at the numbers … I mean, clinicians aren’t just going to materialize out of nowhere. I mean, the projections compared to the population needs are very, very direct. And we’ve got to look at some of these changes.
Bill’s Favorite Transformative Ideas in Healthcare
John Farkas:
So I know that as a part of This Week Health and that community, you work with innovators in this space who are doing great things for transformation. And a few of those that you’ve partnered with happen to be our clients, which is some of why we’re here connected today because they’re big fans. And I’m just curious, as you look at what’s coming out right now, what are you most excited about? What are some of the transformative companies, the transformative ideas that are starting to take hold, they’re starting to come forward that are really going to move the needle?
Bill Russell:
I love the work that Artisight’s doing around artificial sight. I mean, it’s in the title, so it’s cameras located throughout the health system that are trained on algorithm, are trained on the images that it’s seeing, and it can detect falls, it can detect a lot of things. It can detect that this room is empty and the bed needs to be made simple kind of stuff. It can detect falls, it can detect the need for turns, bed turns and those kind of things. It can do a lot of things. Anything that you would normally have a person in there and they’d see something and that would trigger an action. It can be automated through Artisight. To me, that’s fascinating. You’re going to see really the health system, the hospital become a working computer, if you will, that there’s going to be inputs that aren’t all just keyboard and mouse.
It’s going to be cameras, it’s going to be ambient listening, it’s going to be all that stuff feeding into a living organism, which is going to be that health system. And it’s not only going to be that one building. It’s going to be across the entire system. I’m hearing more and more health systems step back and look at system-wide initiatives instead of hospital specific. One of the companies that brought us together is, Shore Test, I’m bullish on Shore Test. They really reduce the financial burden of testing with every piece of software you bring into a new health system. You have to do testing and-
John Farkas:
Or should do testing.
Bill Russell:
Or should do testing. And the thing I like about that solution, and the reason I showcase it is they talk about hours. You have these people that are doing this type of work that could be doing something else, and they’re going to save umpteen number of hours for every release. Well, for a lot of these EHRs, we have to take at least two releases a year, if not more. And every time you take a release, you have to test that against your workflows. You have to test that to make sure you’re not going to do any harm essentially. And so I think those kinds of automation tools that are built to be platforms, because it’s not only the EHR they’re testing, they can also do stuff like Workday and some other platforms that are coming down the pike. So it’s having the ability to implement automation and testing around that is important.
The other one I will mention only because of how innovative it is, and I think there’s there we need more and more innovation in this area is security. And the company is [inaudible 00:51:25]. And [inaudible 00:51:26] has a unique way of identifying PHI, personal health information and personally identifiable information, PII and PHI across the entire health system, identify that, where it lives and resides. I identify when it’s moving, when it shouldn’t be moving and that kind of stuff. The reality is we don’t have enough people to monitor this stuff, find this stuff, mitigate this stuff. We’re going to need tools that can actually do that for us and then quarantine that information and then give us the ability to get in front of these breaches and whatnot. And by the way, I could go on and on. There’s a lot of great companies in healthcare right now trying to solve those problems that have been persistent.
The Importance of Quality Leadership
John Farkas:
So if you were to look at just kind of a summary of a company coming to market, trying to bring a solution forward, you’ve mentioned several things here that are important, but if you were to load their cannon with the most important elements that they need to understand coming into this market, engaging with this market, if you were to give the broad overview, what are the must haves, what are the things that they need to make sure they’re doing? Just kind of overview summary of some of what we talked about here and anything else you’d want to add?
Bill Russell:
Yeah. When I think about a startup, the most important thing is their leadership. So when I’m evaluating, I want to see a leadership team. I want to see a cohesive leadership team that understands what it takes to be successful in healthcare. And so if you don’t have somebody on your team that understands healthcare and the unique environment that healthcare is, even organizations that are really large that are coming into healthcare, one of the first things I say to them is, you need to hire healthcare people. It’s just a distinct animal. And look, I do consulting in every industry out there, and I’m telling you, when healthcare says, “We’re different,” they’re different. And so you need to have those people on staff that can sit across from them. And one of the best examples of that is CrowdStrike hires Drex DeFord, former CIO for Scripps, former CIO for Seattle Children’s. And when Drex walks into a room-
John Farkas:
Somebody knows what’s going on.
Bill Russell:
Somebody who knows what’s going on, somebody who can empathize with where they’re at, and when he walks into the room, they see a friend like, “There’s Drex. I want to talk to Drex. How’s Drex doing?” They don’t see vendor, they see-
John Farkas:
Ally.
Bill Russell:
They see an ally. They also see somebody who’s been in the chair, understands, and I don’t know if everybody’s budget really has that kind of room for it, but if your budget has that kind of room, find somebody that doesn’t just talk, the talk, somebody who’s been in an implementation. If you’re in the EHR world, somebody who’s been in the muck of an EHR implementation and really understands it. If you’re implementing AI in oncology, maybe an oncologist might make sense, a forward-leaning oncologist, I don’t know. So understand healthcare. I think one of the biggest things for me is understanding the distinction between marketing and sales.
People will either come with one bent or the other, and both are incredibly important and they’re distinct. Marketing is about putting your message together, getting your story together, making sure that that story resonates, making sure that it addresses the pain points of the organizations that you’re trying to tell that story to. It’s about getting that story to be heard, getting as much amplification of that message as possible. You know marketing better than I do, but that’s essentially what I would say marketing is. Sales is relationships. Sales is someone talking to someone saying, “Hey, I heard you on this podcast and it sounds like you have a problem that we might be able to help you out with,” or whatever the approach is. But sales is relationships. Sales is a numbers game.
John Farkas:
It’s knowing. Yeah.
Bill Russell:
It’s a numbers game. It’s people talking to people. And if you’re not going to hire a salesperson, somebody in your leadership team needs to do sales, which means needs to have relationships, make phone calls, talk to people.
John Farkas:
And gain that knowledge that you talked about of the situational awareness for that entity so that you can know how to tailor the message that you’re bringing in a way that is going to catch traction because otherwise you’re in some trouble.
Bill Russell:
Or even get feedback to the message. Sometimes when we take the message out, they say … They’ll give us a little piece of a little nugget of gold and we take it back to the product team and say, “Hey, they just said this,” and there’s like an aha moment of, “Oh yeah, that would be so much better if we had that or did that.” You got to be careful because every client wants to tell you how to redo your product completely. You don’t have enough time, runway or money to do that, but sometimes those nuggets really help you to increase your runway and your longevity and then all the things we talked about before. So having a playbook for somebody implementing, really understanding the people you’re going to need, the people you’re going to need access to within their health system, the resources you’re going to need access to, the amount of time you’re actually asking for.
Sometimes these startups will come in and say, “Yeah, just connect us to this,” and I’ll look at them like, “You just asked me to bypass 10 levels of security and go straight into this thing.” My gosh, that’s like two weeks of you sitting down with these people to determine how you’re going to do that, how you’re going to get that data. Is that data anonymized? Is it not anonymized? You should know all those things before you make a request that somebody just looks at you and goes, “You really don’t know what you’re doing, do you?”
Closing Thoughts
John Farkas:
Yeah. Yep. Great input. Hey, as we conclude here, Bill, what would you want our listeners, knowing that these are early health tech companies, what would you want them to know about you and what you’re doing with This Week Health and what you’re doing with the events that you’re putting on? What would be good for them to know and how might they reach out and get in touch with you?
Bill Russell:
Yeah. Anybody can reach me at Bill at ThisWeekHealth.com, and I give it out on my podcast. So it’s the way people get ahold of me. By the way, if you send an email to me and you’re not in my contacts, you’ll be asked to give a donation to Alex’s Lemonade Stand for Cures for Childhood Cancer. If you’re not going to give at least a dollar to that, don’t send me an email because it’s the charity that I want to support and those kind of things, and I figure it has to be worth at least a dollar to you to-
John Farkas:
There you go. That’s awesome.
Bill Russell:
… talk to me. But hey, we started as a podcast. That was five years ago. We’ve done 1,500 interviews, likely the people you’re trying to call on, we’ve talked to some of them. I would say go there, listen to those. We started these 2/29 events. 2/29, February 29th 2020 was the first reported case of COVID to the CDC, and things changed in healthcare, and we saw a need for the smaller events where CIOs can have conversations with each other. We started, those. Those are sponsor driven, and we’ve had great … In fact, our next four meetings are fully booked with sponsors and CIOs, and they have been really going well.
Look, I’ve got about five, maybe 10 years left in this industry doing what I’m doing. I want to make sure that we move the needle forward in healthcare, whatever that looks like. If it’s really pushing the needle on health and the health journey and really getting healthcare systems to get connected with my personal health journey, whatever that looks like moving forward, if that means helping physicians ease some of the burden and the cognitive load that they’re experiencing with the use of AI, I want to see technology have an impact on how healthcare is delivered to me, to my kids, to my grandkids moving forward in the most positive way possible. So if I can help in any way, if you feel like you can help me get a message out there, by all means, let’s do it.
John Farkas:
Awesome. Well, Bill, thanks again and thank you for the service that you’re providing for this community. It’s really important, and as we’ve said, technology has to be a part of the equation. Meaningful, powerful, effective technology has to be a part of the equation that we have moving forward, and I appreciate the part that you’re playing and helping connect that with the healthcare systems that need it so bad. So Bill Russell, thank you so much for joining us today on Healthcare Market Matrix.
Bill Russell:
Thanks, John.
Transcript (custom)
Introducing Bill Russell
John Farkas:
Well, welcome everybody to Healthcare Market Matrix and I’m really excited about who we have the chance to talk with today. There are people who seek to promote themselves by advancing technology and healthcare and then there are people who seek to move healthcare forward by advancing meaningful technology. And today, I know we have the chance to talk to the latter. Bill Russell needs no introduction for many, but for the few of you who don’t know, he is the CEO and founder of Health Lyrics and the podcast host of This Week Health.
Bill has served on lots of executive teams in healthcare, higher education and Fortune 500 consulting practices. He’s the former CIO for St. Joseph’s Health, overseeing a 6.5 billion dollar system with 16 hospitals. He has been in the situation to accelerate IT adoption and strategies and methods including agile cloud development and has successfully invested in a number of startups driving lots of innovation in the healthcare sector and is currently involved in executive coaching and advisory services. He’s providing technology and future focus coaching to a number of health leaders and offers advisory services to the healthcare providers and tech startups and investors through Health Lyrics, which I know we’ll talk a little bit more about. Bill, welcome to Healthcare Market Matrix.
Bill Russell:
John, thanks for inviting me. And anytime you hear your introduction, it makes you feel old. And listening to my introduction, I’m like, “Wow, it seems like forever ago that I was a CIO for St. Joe’s,” and some of the other stuff you didn’t mention. It is 20 plus years in consulting prior to that. And over time, you just get a different perspective on things, I believe.
John Farkas:
Well, and that’s part of why you’re here because you’ve lived to tell the tales. There’s been a lot in that tenure. But really grateful for you joining us here today. And I know that you’ve been in a number, like we mentioned, a number of leadership positions over your career. But as we’ve heard you talk and mentioned a number of times, the CIO role in healthcare has been one of, if not, the hardest. And I’m just curious from your perspective why that is. What do you see as the challenge in that seat?
Bill Russell:
It’s interesting because I’ve had some really hard jobs. I’ve been over a service organization for a Fortune 500 company. Had 2000 engineers reporting it to me in 42 states, I believe, managing 200 million in consulting revenue and that kind of stuff. And somebody would say, “That’s hard.” And then we took that company through chapter 11. “Oh, that’s hard.” And I’m here to tell you they pale in comparison to being a CIO at a health system. And I think the best way to think about it is the CIO role in healthcare is really the CEO of applying technology to that healthcare system. Okay? And what I mean by that is you play every role. Sometimes you set the table and you bring the technology in, sometimes you implement, sometimes you have to rally a group of people around a concept or an idea or around a problem set that needs to be solved.
Sometimes you’re an advisor and counselor. You have to have vision, ask the right amount of questions of people. You have to get the right amount of money to make sure you’re managing your IT organization well. And then of course, then you have all the people challenges, managing people and the various things that go on. And then I think the other thing that adds the complexity because people might say, “That’s not that hard,” and it is hard, but the complexity is healthcare. And when people say the word healthcare to me, I hear a lot of different things. And when a lot of people say the word healthcare, they’re like healthcare provider. Well, to me, a healthcare provider is six types of different organizations. It can be a private equity or venture capital backed organization. It can be an academic medical center and everything in between.
And each one of those is made up of 40 businesses that sometimes are tied together really well and sometimes are not tied together that well and in some cases have competing priorities. And so I set that up. And then I say, when I first went into healthcare as the CIO, it wasn’t uncommon for me to have a meeting at 10:00 AM to talk about surgery. At noon, I’m talking about oncology. At one o’clock, talking about supply chain. Two o’clock, I’m meeting with the vendor and talking enterprise architecture. Three o’clock, I’m talking food services because we have food services in every one of our hospitals. And at 4:00 PM, I’m sitting down with a CFO for a multi-billion dollar organization and we’re talking budget and finance. So hopefully, I painted the picture for why I believe it’s one of the hardest roles, period, out there.
John Farkas:
Yeah. And needing to have some level of mastery and operational understanding of what goes into each of those, right? Because you’re looking to support them and understand the issues and challenges faced by each one of those segments.
Bill Russell:
Yeah, I joke around that when I first went into healthcare, I came from a technology background. I didn’t come from a healthcare background because that’s what our health system wanted. They wanted some new thinking, but I had to get my admin put together a list of all the different practices because I was sitting across from a doctor of some kind and they would say, “Well, I practiced this medicine,” and I would’ve no idea what that was. And so for the first couple of weeks, I had cards and I memorized all the different practices within the health system just to know who I was talking to and what they did.
Tips for Pitching to CIOs
John Farkas:
So good lead in to one of the questions I have here, and we have a number, but I’m just curious. Our audience and who’s listening to this podcast are marketing leaders of health tech organizations. So if you just heard what Bill said and the CIO is a target, somebody you’re trying to get the attention of, that puts forward a pretty significant mandate because first of all, you are working as the CIO to balance an incredibly complex budget with lots of competing demands and needing to make decisions based on clear value. And so if you were to coach a marketing leader of a technology solution selling into a health system, how would you help them get their message across? What would you say is the thing that they need to make sure they’re doing a great job of, if they’re going to get the attention of a CIO and help them understand what it is they’re bringing?
Bill Russell:
It’s so important to know each health system and what their priorities are. It’s interesting, when I started the podcast, we would interview CIOs and there was a few sales managers that told me your show is required listening for our sales organization. I was like, “Really? Tell me about that.” They’re like, “Look, you’re having the conversations that many of us can’t have and you’re asking the questions that many of us can’t ask of the CIOs.” And by the way, I’m not the only one doing it now. You can go to Becker’s, you can go to Health System CIO, a bunch of us interview the CIOs and you can get the priorities from them. So the first thing is understand their priorities. The second is you have to have a story and your story has to have a … What’s the best way to say it? A demonstrable return on not only finance. We all know that. All needs a financial ROI, but it needs a demonstrable return on the resource in investment of moving something forward.
Because, by the way, finance and resources aren’t the only two things I’m going to invest in your project. The other is I’m going to invest my relational capital, my political capital that I’ve built up. If I’m getting behind a project, I’m essentially investing whatever relational and political capital. Because I’m going to have to pull in some doctors and get them excited about it or pull in some administrators and get them excited about it. So I’ve built up some capital. I’m spending that capital when I bring you in, you have to have a story that helps me to make that case internally. No matter what it is. And it has to be clear. It’s amazing to me the number of times where I’m sitting across from somebody and they’re telling me, “You have to implement our system. It’s the blah, blah, blah.”
And I say, “Well, tell me what the return is from a financial standpoint,” and it’s murky at best. And I’m like, “All right, you should have that down.” In fact, in today’s economy and financial situations for health systems today, that has to be rock solid. You have to be able to say, “We can save you.” And it can be in actual dollars, can be in reimbursements, could be in revenue capture, it can be in hours. A lot of times now we’re talking inefficiency, you’re saving us hours. It can be in any of those things, but that story has to be very solid. “You’re going to invest this and you’re going to get this return.” But the other thing is really understanding the problems. So nurse staffing shortages and nurse staffing, just clinical burnout in general. These are real problems. What is your story to go against that and how is that delivering.
The financial burden that I mentioned earlier, what is your story and how are you participating in that? Security, PHI, PII vulnerabilities, how are you participating in that? That’s just to name a few and just understand, there’s an awful lot of things going on in the health system because I’ve had salespeople look at me and go, “Hey, you know what? Bill, I don’t understand why this health system’s not doing this.” And I’m like, “Let me tell you what some of their priorities are.” And I’ll rattle them off and I’ll say, “They have to address sepsis.” And they’ll go, “What is sepsis?” I’m like, “Well, it’s leading cause of death in a hospital. So they have to address that. They have to improve quality.”
They have a staffing issue that’s going on. They have unrest amongst their clinicians, they have burnout issues and those kind of things. They just had to do a reduction in their IT staff. Every one of their vendors is coming to them with a 10 to 15% increase in their price for ongoing software contracts. And you’re asking me the question, are you really asking me the question, why aren’t they prioritizing the stuff that you’re talking about? And the answer is, you might be number 10 on their list and they’re not going to get to number 10 this year.
John Farkas:
Yeah, it’s a really good point. That clarity to value has to be there. It is so frequent that we see organizations wanting to talk all about their technology in some way. Here’s our really cool shiny object that does X, Y, Z, which is great. And if you don’t have a really clear line to value, it means virtually nothing. And that’s a such an important element.
Bill Russell:
By the way, help me sell it internally. If I have to build out all this stuff myself, I don’t have the time and it’s going to be really hard. I want you to hand me this stuff and say, “Hey, if you’re talking to a clinician, here’s the story, take this story to them.”
John Farkas:
Yeah, that’s a great point. Have the internal campaign framework set out so that you can help your advocate bring it forward. That’s really good.
What Problems Are CIOs Solving Today?
John Farkas:
So you outlined several of the problems or you listed several of them, but what are some of the few top problems you’re seeing come across CIO’s desks right now that they’re willing to or feeling urgent about dedicating some resources to solving?
Bill Russell:
The number one problem, and you may not hear this from CIOs because none of them want to really talk about this overtly because it’s a difficult thing, but the number one problem is the financial hole that healthcare is currently trying to climb out of. And we’re seeing some of the leaders climb out of it. You’re seeing Intermountain and Mayo and some of the others climb out of it, but some of the others are really struggling to get out of that. That’s the number one thing that’s happening. We come through the pandemic, which was a jolt to the system for sure. Now we have inflation, we have a sluggish economy, which is never good for healthcare. And oh, by the way, while that’s all happening, the competition has only grown for imaging, for a lot of the things where we used to make money-
John Farkas:
That were revenue sources.
Bill Russell:
… for primary care. So that’s the environment. So that’s number one. So they’re looking for efficiency and potentially revenue generation, so technology that helps in generating revenue, any kind of tech, this is why we’re talking so much about AI right now because there’s the belief that AI can start to step in and take some of these mundane, but some of the mundane work that we all do including in IT and in the clinical side and start to automate some of that. I would say it goes beyond that too though. I mean, once you get past the financial access is always something that people are looking for.
How do we give people more access to healthcare, remote patient monitoring, patient flow? I’m hearing a lot of command centers be stood up because we would have people get stuck in the system and instead of moving efficiently through, they would get stuck. And even to the point of discharge, we have some patients who are sitting maybe four or five hours taking up a room in a hospital where they could have been discharged five hours ago. And I mean, all those things represent assets and I think that the last one I would mention is digital expectations have changed and it’s due to all the things that people talk about, it’s due to the Amazon effect or-
John Farkas:
And so when you say digital expectations, you’re saying how can people engage digitally in whatever framework?
Bill Russell:
Right. You hear this all the time and maybe you’ve experienced it yourself, which is, “Hey, why can’t I get an appointment through my phone? Why can’t I cancel it? Why can’t I reschedule it? Why can’t I get my discharge instructions via some digital tool that I can then show to people on my TV at home?” Our expectations have changed because every other industry has changed them because we’re already in retail, we’re there, everywhere else, we’re there. And it’s almost cliche to talk about now what we’re pushing the CIOs in terms of the conversation is if those are their expectations, what are the projects we’re putting in place to fulfill the expectations that they do have in terms of how healthcare will operate over the next five years?
Finding New Tech Solutions through Other Health Systems
John Farkas:
Yeah, that makes a lot of sense. So thinking back at your days at St. Joseph, if you were looking to or how were different vendors, different technology organizations, how are they coming across your desk? How are you becoming aware of them? What were some of the channels and methods that you were hearing of new opportunities?
Bill Russell:
I ended up writing an article about this. The number one way that people hear about solutions is from other health systems period. And it’s like a three to one, four to one, five to one margin over whatever number two is. Because we just had one of these events and the CIOs were sitting around and somebody said, “Hey, we were able to get more efficiency out of our MRI machines by utilizing this thing.” And three people swarmed that person after that presentation to say, “Hey, I needed more information on that.” That’s a very common approach. You’ll see that happen at conferences. You’ll see people say, “Hey, we’re kicking off a project here,” and we think they’re going to start at Gartner, or we think or they’re going to start a class or one of those other services. But a lot of times what they do is they’ll essentially go out into the field.
There are groups of people where they can send an email to and say, “Hey, who’s doing this specific thing for hospital at home with remote patient monitoring?” And they’ll get two or three responses. They’ll get their teams on the phone with them and that’s how they’ll learn. That’s number one. Every now and then, I would do the booth crawl at various conferences, but most of the time when I go to do the booth crawl, I’m looking for my existing partners so that I can have conversations with them and whatnot. Rarely was I doing the booth crawl to find a new solution. Every now and then I’d have a doctor with me and the doctor would say, “Hey, I want you to see this thing, or I want you to take a look at this thing.” That would happen.
And then there’s a story now that has sort of grown legend in that one of the people who called on me used to be with Salesforce and he now does sales coaching and he has this whole section where he talks about effectively marketing to a health system and I’m the muse in this whole thing. He sent me emails day in, day out. It felt like that. He said he sent me seven emails before he got a response and it led to a seven figure deal for him and for Salesforce when he finally got through to me. But again, that requires such persistence and he was really good at identifying the things that I was thinking about and just saying, “Hey, I heard this. I heard you do an interview, I heard you speak.” And he would drop these things and eventually, he got in touch with me. We got into a room and he listened and found a way to help us.
John Farkas:
So some of what I’m hearing there that I think is really important is, first of all, you have to know what’s going on in the organization. You have to know what’s top of mind. You have to understand what are the core problems that they’re facing in trying to solve and tailor whatever approach you might be taking with some very clear understanding of the context. Because if you come out in blind and you’re just throwing a blanket solution or a blanket approach, it’s likely not going to get the attention. The other thing is recognizing that peer to peer is going to be some of the most effective channels. So figuring out how to use existing deployments, existing successful deployments and amplified those out so people can see that and understand what in fact is going on, what in fact is being realized as far as value goes in a real respectable system that people have connection to. That seems like those are some important takeaways.
Bill on Failed Implementations
Bill Russell:
Yeah. You cannot afford to have a failed implementation. It’s too small. It’s too small of a community to have failed implementations. If you do, I would go back and remedy those as quickly as you possibly can.
John Farkas:
Yeah, talk about what you’ve experienced in that regard, because I’ve definitely heard, especially recently and especially around the AI universe of some things not going well and that spreading pretty rampantly through the rumor mill.
Bill Russell:
Yeah, I mean, I’ll speak to the rumor mill. The rumor mill is pretty obvious. It’s one of those where you’re sitting with a group of CIOs and we get into that old curmudgeonly kind of thing. “You think your life’s bad. Let me tell you how bad mine is. I went to implement X, Y, Z technology. Oh my gosh, these people were idiots. They absolutely didn’t know what they were doing and blah, blah.” Even if that company did know what they were doing or whatever, that person who’s relaying that story isn’t going to say it that way. They’re going to tell you all the things that went wrong. They’re going to embellish in some cases. I mean, you cannot afford, it just happens. I mean, I’ve just seen it, I’ve seen it, I’ve heard it over and over again. So you have to make sure you remedy those as quickly as you can.
At the minimum, you want to get people to be neutral on you if you have a failed implementation because we know. In healthcare, we know that a lot of times we’re the problem. It took us six months to get the contract together. We didn’t get you the data correctly. We didn’t get you the integrations correctly. Our doctors didn’t engage. There’s any number of reasons the project didn’t go well that are literally our fault. But if it’s going to fail, at least get that neutral feeling of, “Hey, let’s pull the plug on this. It doesn’t seem like this is a priority for you right now. Let’s get to a point where it’s stable and we can cut our losses or whatever.” That’s the minimum. At most, I would tell organizations, “Man, you have to be really good at managing your clients through a project. You should know what it takes, who needs to be involved.”
In fact, I was so impressed when I had a solution and they came in and said, “These are the people that we are going to need to engage with. This is the amount of time we’re going to need from those people. This is the connections we’re going to need. Here’s the security framework that we’re going to utilize.” They knew the questions we were going to ask before we asked them and they provided them complete with, “Here’s a template for the project management.” A lot of them will provide former healthcare staff that have been through it before that can help us to walk through.
John Farkas:
We can walk alongside. Yeah.
Bill Russell:
All that stuff is really important. And by the way, I mean, EPIC has been successful really for a lot of reasons. But the two early on I would say is, one is they were the only system with an acute and ambulatory integrated system when meaningful use came out. So that’s the number one reason they won is that. But the number two reason they won is when EPIC comes into your healthcare system, they say, “You’re going to get this many doctors trained. You’re going to do this, you’re going to do this. Here are the recipes for success and we’re going to be with you and walk you through it every step of the way.” And healthcare, they like that. They’ve grown used to it. They’ve grown used to it with EPIC and they would like to see it from more players who come in there and really know how to get stuff done in healthcare.
John Farkas:
Which is really the challenge for a lot of early stage health tech companies, right? Because they have a relatively thin team. Sometimes they don’t have that savvy expertise. And so spending the time on the front end to get that, because you can’t afford a failed implementation, especially early on, you have to come … And so spending the calories that you need to go to school and understand what it’s going to take to be successful and being successful at almost any expense, especially in the early days, is a really critical step. Because, first of all, you need the experience and you’re going to continue to need the experience because technology isn’t enough. If you’re a CIO and you’re implementing a six figure or a seven figure, an eight figure system and you can’t afford for it to fail. Your reputation can’t sustain it and your budget can’t sustain it and the system can’t sustain it right now.
Partnering with Health Systems
Bill Russell:
Really early on, I will say it’s so important to get a partner and I tell really early stage startups, “Set aside some ownership for a partner.” So they’re going to partner with me at St. Joe’s. What they should do is offer me warrants. They should offer me some ownership. And the reason they should do that is because I’m taking a flyer on them. They’re so early on and what they need from me is they need some of my doctor’s time to say, “Man, I see what you’re doing here. You shouldn’t do it this way. You should consider doing it this and this because you don’t really understand how this works in healthcare.”
And they might need some help from my security team to shore up their security framework or my integration team to do some things. They want me to feel like I’m an owner so that I’m going to give them as many resources as possible. I’m also going to be in senate to give them a positive review out into the world. So I would say set aside some ownership early on for one or two players and then I would say stop giving away ownership once you have three or four clients. I mean, I see people who are like, “Oh yeah, we gave them a piece of ownership and they’re on client number 10.” I’m like, “What are you doing? Your product should be baked by now.”
John Farkas:
Well, having a very clear principle approach to that onboarding in those … You need to go into those first deployments, understanding that we’ve got to get the system together. That’s our purpose right now is to get our system together so that by the time you get to number four, you’ve got it down and you have the opportunity to push it forward.
Bill Russell:
And hopefully in working with me, I’m going to give you the project plan that we used and then you could use that with other clients.
Bill’s Expectations for AI’s Impact on Healthcare
John Farkas:
Yep. Yep. Good. That’s a great word. So let’s shift gears a little bit. I know you’re the host of This Week Health, we hear you talking a lot with CIOs and in that community. We know that AI is a strong emerging force on a number of fronts. What are you seeing some of the opportunities? If you look at the opportunities that AI is going to open up for healthcare and how it intersects with the CIO universe, what are you seeing as some of the big beachhead there that are going to be transformative in the next few years?
Bill Russell:
AI is going to change everything and it’s going to do so very rapidly. And what I’m telling CIOs right now is stop putting up roadblocks. Stop moving so slowly because this is going to change everything and your competitors are trying to figure out how to use it. One of the CIOs I respect greatly is Mike Pfeffer from Stanford and they’re already using AI to draft notes for doctor’s responses and then once it’s drafted, the doctor can look at it and say, “Go ahead and send it.” And they’re saving a certain amount of time and the studies will continue to be done. But those doctors used to have to write those notes and they actually did a study on the notes for empathy and for completeness, and the AI generated notes were more empathetic and more complete than the ones the doctors were doing. Why is that? Well, the doctor just worked an eight to 10 hour day.
John Farkas:
They don’t have time.
Bill Russell:
Right. They’re at the end of their day. They’re just like, “Yeah, go ahead and take the medication or let me do it.” Whereas AI sitting there going, “Hey, thank you for your question. I really appreciate that. I’ve been looking through your record,” all the stuff that the doctor wouldn’t type. Anyway, I start with saying it’s going to change everything. AI is going to be an information assistant. It’s going to follow me all day. It’s going to document what needs to be documented in my day. It’s going to send emails or at least draft emails for me that need to be sent. It’s going to assist me with tasks that can be done digitally, like ordering something or those kind of things. It’s going to do research on my behalf and it’s going to provide me all sorts of support in my work.
This is what AI has the potential to do. Clinically, this means that every physician will have an AI assistant that can gather the information for their patients that they’re going to see that day, all right? And it’s going to get the latest research from the journals on those patients that they’re going to see. It’s going to score those for relevance for them. It’s going to provide them a summary on some of the bullet points for the patients that they’re going to see. It’s going to listen in on the patient visit and it’s going to document that patient visit. It may even pull up some relevant information from the EHR while they’re having that visit. You get the picture. From a clinician standpoint, the anticipation is this is going to be used everywhere to do things that we used to have to go to our mouse, go to our keyboard and go, “I wonder what the last 10 readings were on blood glucose for this patient.”
And by the way, I’m just talking about clinicians now. Nurses have a lot of, again, tasks that are, depending on their level of certification, some of the tasks that nurses are asked to do are quite frankly beneath their level of certification. And those tasks we have proven now can be done with cameras and computer vision utilizing AI and those kind of things. What we are really cognizant of right now though, is AI washing. Look, we’ve been doing predictive models for years without algorithms. That is not AI. AI is with every piece of information that washes over the AI algorithms that are there, it actually gets more intuitive. It gets smarter. It knows where to find information better. It knows how I function better. It gets smarter as we go along. That’s what we’re looking for is algorithms that get better with use.
John Farkas:
And clearly, one of the things that … what we know right now is one of the biggest needs that is facing healthcare has to do with the clinician shortage and what is going on. Really critical that whatever solution you’re bringing forward has a very clear line to actually helping and lifting the burdens off of clinicians. It’s one thing to throw another technology layer over something and it’s another thing to have a solution that is, from its core, from its foundation, built to save cycles and to improve care quality. And I think that that’s a real critical element as you consider how you are messaging, how you are bringing these ideas forward, how’s that going to be clearly related?
Bill Russell:
Yeah. It goes back to what we talked about earlier. You have to have that demonstrable … I mean, this is why you need that early client as well so that you can write that paper that says, “Hey, we reduced their clinician time in notes or follow up or in documentation by 25%, which it allowed them to do whatever they want with that time.” But in a lot of cases, the clinicians, they say they’re suffering from burnout, but when we give them more time what they do? They see two more patients a day. Go figure.
The Need for AI Policies in Healthcare
John Farkas:
Bill, as you’re interacting with CIOs and in the context of AI, where is the biggest knowledge gap right now? What’s the gap that needs to be closed between the average? I know there’s a huge spectrum of embrace right now around AI ideas and there’s been a whole lot that’s just come out in the last eight months around what is now possible as the large language models have come to roost. What are you seeing as the gap that exists in fundamental understanding right now that needs to get closed that organizations who are bringing AI solutions forward would do well to invest in helping people understand?
Bill Russell:
This is an interesting question and part of the discussion we had this past weekend.I hope people understand, you just can’t pop technology in a place like healthcare. You can do it in some industries. You just can’t do it in healthcare. It’s highly regulated.
John Farkas:
It’s too much context.
Bill Russell:
Right. It’s highly regulated. And the mistake could cost someone their life or the life of a loved one. So either way, you can’t just pop it in there. There’s foundational work that needs to be done before AI can be widely adopted and used. I would start with we have to create policies, and I asked this group of CIOs, how many of you have an AI usage policy in place? And of the 13 that were in the room, two thirds were working on it, but no one had one done.
John Farkas:
Wow. Yeah.
Bill Russell:
I know it didn’t, but it’s like all of a sudden with chatGPT, all of a sudden became, “Oh my gosh, we need an AI policy.” Now we were using AI prior to that, but it wasn’t as pervasive and there wasn’t as much demand to use it. And so we didn’t put the policies around it and we have to put those policies around it. The policies have to protect people, data has to be protected. Obviously, we’re doing AI in supply chain, we’re doing it in IT, areas where there’s no lives at risk, but we have to do that.
And while we’re talking about policy, this week’s poll, I asked … So I do polls every Monday on LinkedIn just to get the voice of the industry. And my question this week was around policy. If you went to a healthcare system and you read their AI policy, what would you be really concerned if you didn’t see on there? And with LinkedIn, you can only give four options. I would’ve given eight if I could, but the four options I gave were data privacy and security, transparency and explainability, fairness and bias mitigation and human oversight and accountability. These are just four of the headings. I think that should be in every AI policy.
Data privacy and security, transparency, explainability, we can explain why the AI algorithm is telling us what it’s telling us. Fairness and bias mitigation because we know that these algorithms were trained and they’re trained on data from the internet. And that isn’t necessarily bias free information. And the example I give people is we don’t know if chatGPT has been trained on CNN or FOX News and if it was trained on one or the other, it’s going to answer the question different. And so fairness and bias mitigation or human oversight and accountability.
And right now within healthcare, I see in what I’m recommending for people is the human needs to be still the connection to the patient. AI can’t automatically do anything yet. It can reveal information, it can bring it up to the forefront, it can draft things, but the person who actually read it is a human. It’s a physician, it’s a nurse, it’s somebody who can say, “Because I’m pressing this button-“
John Farkas:
I’m responsible.
Bill Russell:
“… I’m responsible for this.” Let me give you the results so far. Data privacy and security, 44% people said, “Hey, if I don’t see that in the policy, I’m going to be concerned.” And 40% said human oversight and accountability. So there’s 84% right there think that those are probably the top two topics that need to be addressed in any AI policy right now.
John Farkas:
Very, very insightful. I know we’re working with a group right now that is very actively working to promote the importance and the methodology for doing an audit of large language models to help ensure if you’re deploying any form a large language model, you better know what’s in there and you better know where it’s coming from and be able to look at what’s going on. And that’s going to be an increasingly important element as this stuff comes forward. And the whole privacy framework around it obviously is essential. What do you see anybody currently doing work on helping promote guidance for those policies other than yourself
Bill Russell:
And I’m. I haven’t had a consulting client in six months. I’m so busy with the stuff that I am doing, building the community and stuff over here. I will give you one funny answer. I said, “How many of you written a policy?” And one of the CIOs said, “Oddly enough, we asked chatGPT to write a AI governance policy for healthcare. And it wasn’t half bad. I mean, it was pretty good actually,” is what the person said. And I’m like, “That’s interesting. AI writing a policy, our large language model writing a policy to oversee AI and implementations.” But quite frankly, based on my usage of chatGPT and GPT 4, it probably would write a fairly good policy statement. Who’s advising on this? At my health system, it would’ve been compliance. We would’ve had physicians. We would’ve had our legal team. Quite frankly, I’m recommending that everybody set up an AI governance board. It’s going to be pervasive in healthcare over the next five years.
It’ll be slow to get going but it will pick up steam very rapidly. And the same way you have governance around data and governance around applications, you’re going to need AI governance because you said transparency in the model. There’s no transparency in the chatGPT model. Nobody can tell you what it’s been trained on. I mean, they can generally tell you what it’s been trained on, but outside of that, we don’t really know what it’s been trained on. And so if it doesn’t have that level of transparency, should it be used in healthcare? Some people are saying yes, as long as it has human oversight, but my encouragement to any healthcare system listening to this is the same way you have an ethics board and those kind of things, you have to determine how this is going to be used in your system and what’s appropriate. Keep in mind if you block it and slow it down, you might be signing your own foreclosure notice. I mean, because it can fundamentally change the economics of healthcare. And if it does that and you decide to step back, don’t be surprised if the financial-
John Farkas:
Don’t be surprised if it bites you.
Bill Russell:
Right.
AI’s Impact on Other Industries
John Farkas:
Yeah, that makes sense. Well, that’s definitely good input. I know that you recently shared an article on LinkedIn where Scott Galloway shares how the economic impact of AI won’t be the catastrophe that many predict. Where do you see that? Where do you see the biggest improvement? Where do you see the biggest impact on AI?
Bill Russell:
We’ve already talked about somewhat in healthcare where I see it, if I were to say where we’re going to see the biggest impact upfront the next two years, we’re going to see it in finance. Imagine AI models essentially managing your portfolio. Now, instead of trusting that financial planner and whatnot, you’re really trusting an algorithm that is looking at way more data points than that flashy person who sold you on, “Trust me with your money, I’ll take care of you.” I think we’re already seeing a lot of AI models in finance. Finance will probably lead the way in this retail will be a close second. When I say close second, I mean, hours behind finance in terms of the implementation of AI. We’re going to see AI really hyper personalizing our experience in retail in a way that is almost unimaginable right now.
And then what will happen, and I had this conversation with Taylor Davis who originally put me onto this concept. Taylor Davis was with class and he led the arch collaborative for many years for class. He was a VP over there and he just started his own deal. And when we were talking about this, he said, “We’ll see industries literally change in what will feel like overnight over the next couple of years. And then healthcare will look at it and go, ‘Oh my gosh, this really is fundamentally going to change things.’ And then we will start adopting it.” So it still might feel really slow.
This thing started, I’m saying the start date was GPT 3 and 3.5 just before Christmas of last year. Just before Christmas of 2022 is around the timeframe that this started. I will say that’s when the clock started. Within three years, we’ll be looking at this across the board in healthcare. It’ll be in supply chain, it’ll be in revenue cycle, it’ll be in IT. It’ll be pervasive in those areas. Administration, it’ll be pervasive in that area. And then you’ll start to really see it take off in the clinical side by then. And it’ll be pretty well expected. A doctor coming out of med school will be like, “You don’t have any AI tools that are going to help me be a better physician? I’m going to go to a different health system.”
John Farkas:
Gotcha. Yeah, that makes total sense. It is going to be a day in this next little bit and how we embrace the waves that are coming is going to make or break a lot of these organizations, for sure.
Why Healthcare Needs to Integrate Quickly
Bill Russell:
John, can I give you my disclaimer? My disclaimer is in all the years I’ve been in healthcare, which is now a little over a decade, all the changes that I thought would be two years away have been four. And the ones I thought would be three years away have been six. We moved to the cloud in 2012 at St. Joe’s, 2012, 2013, and I’m like, “Everybody’s going to move to the cloud.” When they’re moving to the cloud, they just are now moving to the cloud. It’s a decade later from when I thought. I can’t believe this is going to be widespread adoption. It took almost a decade.
John Farkas:
Only a decade.
Bill Russell:
Just moves a little slower. It’s a little …
John Farkas:
What I’d like to hope and is that the economic incentives right now are so undeniable that the innovation may come quicker. I hope they can see their way.
Bill Russell:
That’s why I’m saying that. I do believe that the economics of healthcare are creating a sense of urgency. I believe that the labor situation in healthcare is creating a sense of urgency. I’ve heard people say we will never have enough doctors or nurses again in our lifetime. And so if that’s going to be the case, and that is … We’re not trying to put doctors or nurses out of work. We’re trying to make sure we have enough care knowledge and caregiving being done. And that can be done through a really smart chatbot that’s been trained on the right information and with the right oversight and the right transparency around that model. There’s a belief that 50% of all primary care could potentially be at least triaged if not provided, outright provided by AI within the next five years.
John Farkas:
Yeah. And that scares a lot of people. And if you look at the numbers, it’s almost necessitated, right? I mean, it’s got to change and we’ve got to be open to new ways of addressing some of these things and it’s a big wide open door for really smart technology. But boy, it has to be well-founded.
Bill Russell:
And John, you were worries I wasn’t going to say anything provocative.
John Farkas:
There you go. I think that you jumped right out there, right in it. But anybody who’s looked at the numbers … I mean, clinicians aren’t just going to materialize out of nowhere. I mean, the projections compared to the population needs are very, very direct. And we’ve got to look at some of these changes.
Bill’s Favorite Transformative Ideas in Healthcare
John Farkas:
So I know that as a part of This Week Health and that community, you work with innovators in this space who are doing great things for transformation. And a few of those that you’ve partnered with happen to be our clients, which is some of why we’re here connected today because they’re big fans. And I’m just curious, as you look at what’s coming out right now, what are you most excited about? What are some of the transformative companies, the transformative ideas that are starting to take hold, they’re starting to come forward that are really going to move the needle?
Bill Russell:
I love the work that Artisight’s doing around artificial sight. I mean, it’s in the title, so it’s cameras located throughout the health system that are trained on algorithm, are trained on the images that it’s seeing, and it can detect falls, it can detect a lot of things. It can detect that this room is empty and the bed needs to be made simple kind of stuff. It can detect falls, it can detect the need for turns, bed turns and those kind of things. It can do a lot of things. Anything that you would normally have a person in there and they’d see something and that would trigger an action. It can be automated through Artisight. To me, that’s fascinating. You’re going to see really the health system, the hospital become a working computer, if you will, that there’s going to be inputs that aren’t all just keyboard and mouse.
It’s going to be cameras, it’s going to be ambient listening, it’s going to be all that stuff feeding into a living organism, which is going to be that health system. And it’s not only going to be that one building. It’s going to be across the entire system. I’m hearing more and more health systems step back and look at system-wide initiatives instead of hospital specific. One of the companies that brought us together is, Shore Test, I’m bullish on Shore Test. They really reduce the financial burden of testing with every piece of software you bring into a new health system. You have to do testing and-
John Farkas:
Or should do testing.
Bill Russell:
Or should do testing. And the thing I like about that solution, and the reason I showcase it is they talk about hours. You have these people that are doing this type of work that could be doing something else, and they’re going to save umpteen number of hours for every release. Well, for a lot of these EHRs, we have to take at least two releases a year, if not more. And every time you take a release, you have to test that against your workflows. You have to test that to make sure you’re not going to do any harm essentially. And so I think those kinds of automation tools that are built to be platforms, because it’s not only the EHR they’re testing, they can also do stuff like Workday and some other platforms that are coming down the pike. So it’s having the ability to implement automation and testing around that is important.
The other one I will mention only because of how innovative it is, and I think there’s there we need more and more innovation in this area is security. And the company is [inaudible 00:51:25]. And [inaudible 00:51:26] has a unique way of identifying PHI, personal health information and personally identifiable information, PII and PHI across the entire health system, identify that, where it lives and resides. I identify when it’s moving, when it shouldn’t be moving and that kind of stuff. The reality is we don’t have enough people to monitor this stuff, find this stuff, mitigate this stuff. We’re going to need tools that can actually do that for us and then quarantine that information and then give us the ability to get in front of these breaches and whatnot. And by the way, I could go on and on. There’s a lot of great companies in healthcare right now trying to solve those problems that have been persistent.
The Importance of Quality Leadership
John Farkas:
So if you were to look at just kind of a summary of a company coming to market, trying to bring a solution forward, you’ve mentioned several things here that are important, but if you were to load their cannon with the most important elements that they need to understand coming into this market, engaging with this market, if you were to give the broad overview, what are the must haves, what are the things that they need to make sure they’re doing? Just kind of overview summary of some of what we talked about here and anything else you’d want to add?
Bill Russell:
Yeah. When I think about a startup, the most important thing is their leadership. So when I’m evaluating, I want to see a leadership team. I want to see a cohesive leadership team that understands what it takes to be successful in healthcare. And so if you don’t have somebody on your team that understands healthcare and the unique environment that healthcare is, even organizations that are really large that are coming into healthcare, one of the first things I say to them is, you need to hire healthcare people. It’s just a distinct animal. And look, I do consulting in every industry out there, and I’m telling you, when healthcare says, “We’re different,” they’re different. And so you need to have those people on staff that can sit across from them. And one of the best examples of that is CrowdStrike hires Drex DeFord, former CIO for Scripps, former CIO for Seattle Children’s. And when Drex walks into a room-
John Farkas:
Somebody knows what’s going on.
Bill Russell:
Somebody who knows what’s going on, somebody who can empathize with where they’re at, and when he walks into the room, they see a friend like, “There’s Drex. I want to talk to Drex. How’s Drex doing?” They don’t see vendor, they see-
John Farkas:
Ally.
Bill Russell:
They see an ally. They also see somebody who’s been in the chair, understands, and I don’t know if everybody’s budget really has that kind of room for it, but if your budget has that kind of room, find somebody that doesn’t just talk, the talk, somebody who’s been in an implementation. If you’re in the EHR world, somebody who’s been in the muck of an EHR implementation and really understands it. If you’re implementing AI in oncology, maybe an oncologist might make sense, a forward-leaning oncologist, I don’t know. So understand healthcare. I think one of the biggest things for me is understanding the distinction between marketing and sales.
People will either come with one bent or the other, and both are incredibly important and they’re distinct. Marketing is about putting your message together, getting your story together, making sure that that story resonates, making sure that it addresses the pain points of the organizations that you’re trying to tell that story to. It’s about getting that story to be heard, getting as much amplification of that message as possible. You know marketing better than I do, but that’s essentially what I would say marketing is. Sales is relationships. Sales is someone talking to someone saying, “Hey, I heard you on this podcast and it sounds like you have a problem that we might be able to help you out with,” or whatever the approach is. But sales is relationships. Sales is a numbers game.
John Farkas:
It’s knowing. Yeah.
Bill Russell:
It’s a numbers game. It’s people talking to people. And if you’re not going to hire a salesperson, somebody in your leadership team needs to do sales, which means needs to have relationships, make phone calls, talk to people.
John Farkas:
And gain that knowledge that you talked about of the situational awareness for that entity so that you can know how to tailor the message that you’re bringing in a way that is going to catch traction because otherwise you’re in some trouble.
Bill Russell:
Or even get feedback to the message. Sometimes when we take the message out, they say … They’ll give us a little piece of a little nugget of gold and we take it back to the product team and say, “Hey, they just said this,” and there’s like an aha moment of, “Oh yeah, that would be so much better if we had that or did that.” You got to be careful because every client wants to tell you how to redo your product completely. You don’t have enough time, runway or money to do that, but sometimes those nuggets really help you to increase your runway and your longevity and then all the things we talked about before. So having a playbook for somebody implementing, really understanding the people you’re going to need, the people you’re going to need access to within their health system, the resources you’re going to need access to, the amount of time you’re actually asking for.
Sometimes these startups will come in and say, “Yeah, just connect us to this,” and I’ll look at them like, “You just asked me to bypass 10 levels of security and go straight into this thing.” My gosh, that’s like two weeks of you sitting down with these people to determine how you’re going to do that, how you’re going to get that data. Is that data anonymized? Is it not anonymized? You should know all those things before you make a request that somebody just looks at you and goes, “You really don’t know what you’re doing, do you?”
Closing Thoughts
John Farkas:
Yeah. Yep. Great input. Hey, as we conclude here, Bill, what would you want our listeners, knowing that these are early health tech companies, what would you want them to know about you and what you’re doing with This Week Health and what you’re doing with the events that you’re putting on? What would be good for them to know and how might they reach out and get in touch with you?
Bill Russell:
Yeah. Anybody can reach me at Bill at ThisWeekHealth.com, and I give it out on my podcast. So it’s the way people get ahold of me. By the way, if you send an email to me and you’re not in my contacts, you’ll be asked to give a donation to Alex’s Lemonade Stand for Cures for Childhood Cancer. If you’re not going to give at least a dollar to that, don’t send me an email because it’s the charity that I want to support and those kind of things, and I figure it has to be worth at least a dollar to you to-
John Farkas:
There you go. That’s awesome.
Bill Russell:
… talk to me. But hey, we started as a podcast. That was five years ago. We’ve done 1,500 interviews, likely the people you’re trying to call on, we’ve talked to some of them. I would say go there, listen to those. We started these 2/29 events. 2/29, February 29th 2020 was the first reported case of COVID to the CDC, and things changed in healthcare, and we saw a need for the smaller events where CIOs can have conversations with each other. We started, those. Those are sponsor driven, and we’ve had great … In fact, our next four meetings are fully booked with sponsors and CIOs, and they have been really going well.
Look, I’ve got about five, maybe 10 years left in this industry doing what I’m doing. I want to make sure that we move the needle forward in healthcare, whatever that looks like. If it’s really pushing the needle on health and the health journey and really getting healthcare systems to get connected with my personal health journey, whatever that looks like moving forward, if that means helping physicians ease some of the burden and the cognitive load that they’re experiencing with the use of AI, I want to see technology have an impact on how healthcare is delivered to me, to my kids, to my grandkids moving forward in the most positive way possible. So if I can help in any way, if you feel like you can help me get a message out there, by all means, let’s do it.
John Farkas:
Awesome. Well, Bill, thanks again and thank you for the service that you’re providing for this community. It’s really important, and as we’ve said, technology has to be a part of the equation. Meaningful, powerful, effective technology has to be a part of the equation that we have moving forward, and I appreciate the part that you’re playing and helping connect that with the healthcare systems that need it so bad. So Bill Russell, thank you so much for joining us today on Healthcare Market Matrix.
Bill Russell:
Thanks, John.
About Bill Russell
Bill Russell has served on executive teams in healthcare, higher education and Fortune 500 consulting practices.
As Chief Information Officer for St. Joseph Health, a 16-hospital $6.5 billion system, he rapidly accelerated the diffusion of new IT strategies and methods while also enhancing organization stability and improved IT performance. This was accomplished by adopting emerging models like cloud, agile development and investing in two successful startups.
Bill provides executive coaching to health leaders with an emphasis on technology and futures. In addition he provides advisory services to healthcare providers, health tech startups, and investors through Health Lyrics.
On This Week in Health IT, Bill utilizes his experience to prompt leaders, innovators and organizations to share their stories in a conversational style to capture the wisdom of the industry.